Diallyl trisulfide-induced apoptosis in human prostate cancer cells involves c-Jun N-terminal kinase and extracellular-signal regulated kinase-mediated phosphorylation of Bcl-2

Abstract

Garlic-derived organosulfides (OSCs) including diallyl trisulfide (DATS) are highly effective in affording protection against chemically induced cancer in animals. Evidence is also mounting to indicate that some naturally occurring OSCs can suppress proliferation of cancer cells by causing apoptosis, but the sequence of events leading to proapoptotic effect of OSCs is poorly defined. Using PC-3 and DU145 human prostate cancer cells as a model, we now demonstrate that DATS is a significantly more potent apoptosis inducer than diallyl sulfide (DAS) or diallyl disulfide (DADS). DATS-induced apoptosis in PC-3 cells was associated with phosphorylation of Bcl-2, reduced Bcl-2 : Bax interaction, and cleavage of procaspase-9 and -3. Bcl-2 overexpressing PC-3 cells were significantly more resistant to apoptosis induction by DATS compared with vector-transfected control cells. DATS treatment resulted in activation of extracellular-signal regulated kinase 1/2 (ERK1/2) and c-jun N-terminal kinase 1 (JNK1) and/or JNK2, but not p38 mitogen-activated protein kinase. Phosphorylation of Bcl-2 in DATS-treated PC-3 cells was fully blocked in the presence of JNK-specific inhibitor SP600125. Moreover, JNK inhibitor afforded significant protection against DATS-induced apoptosis in both cells. DATS-induced Bcl-2 phosphorylation and apoptosis were partially attenuated by pharmacological inhibition of ERK1/2 using PD98059 or U0126. Overexpression of catalase inhibited DATS-mediated activation of JNK1/2, but not ERK1/2, and apoptosis induction in DU145 cells suggesting involvement of hydrogen peroxide as a second messenger in DATS-induced apoptosis. In conclusion, our data point towards important roles for Bcl-2, JNK and ERK in DATS-induced apoptosis in human prostate cancer cells.

Introduction

Medicinal benefits of the vegetables of genus Allium, such as garlic and onions, are well documented and include lipid-lowering and cardiovascular effects and anticancer effects (Dausch and Nixon, 1990; Agarwal, 1996). Evidence for anticancer activity of Allium vegetables derives from epidemiological data as well as laboratory studies. Epidemiological studies have documented an inverse correlation between dietary intake of Allium vegetables and the risk for certain types of cancers including prostate cancer (You et al., 1989; Hsing et al., 2002). Laboratory studies indicate that the anticarcinogenic activity of Allium vegetables is due to organosulfur compounds (OSCs) that are released upon processing (cutting or chewing) of these vegetables (Dausch and Nixon, 1990; Dorant et al., 1993).

Garlic-derived OSCs, including diallyl sulfide (DAS), diallyl disulfide (DADS) and diallyl trisulfide (DATS), have been shown to offer significant protection against cancer in animal models induced by a variety of chemical carcinogens (Wargovich, 1987; Sparnins et al., 1988; Wargovich et al., 1988; Wattenberg et al., 1989; Sumiyoshi and Wargovich, 1990; Takahashi et al., 1992; Reddy et al., 1993; Schaffer et al., 1996; Suzui et al., 1997). For example, N-nitrosodiethylamine-induced neoplasia of the forestomach in mice was inhibited significantly by DADS administration (Wattenberg et al., 1989). Intragastric intubation of DADS provided protection against colon and renal neoplasia in a multiorgan carcinogenesis model in male F344 rats (Takahashi et al., 1992). Cancer chemoprevention by naturally occurring OSC analogues has also been observed against N-nitrosomethylbenzylamine-induced esophageal cancer in rats (Wargovich et al., 1988), dimethylhydrazine-induced colon cancer in mice (Wargovich, 1987; Sumiyoshi and Wargovich, 1990), benzo[a]pyrene-induced forestomach and pulmonary carcinogenesis in mice (Sparnins et al., 1988), and N-methyl-N-nitrosourea and 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine-induced mammary cancer in rats (Schaffer et al., 1996; Suzui et al., 1997).

More recent studies have shown that some OSC analogues can inhibit proliferation of cultured cancer cells by causing apoptosis and/or cell cycle arrest (Sundaram and Milner, 1996; Knowles and Milner, 1998, 2001; Nakagawa et al., 2001; Robert et al., 2001; Shirin et al., 2001; Kwon et al., 2002; Filomeni et al., 2003; Xiao et al., 2003a). Milner and co-workers were the first to report apoptosis induction by DADS in HCT-15 human colon cancer cells, which correlated positively with an increase in intracellular free calcium levels (Sundaram and Milner, 1996). In MDA-MB-231 human breast cancer cell line, the DADS-induced apoptosis was associated with upregulation of Bax, downregulation of Bcl-XL and activation of caspase-3 (Nakagawa et al., 2001). Induction of apoptosis and an increase in caspase-3-like activity was also observed in SW-480 human colon cancer cell line following treatment with a water-soluble OSC analogue S-allyl mercaptocysteine (Shirin et al., 2001). Even though these studies have provided convincing evidence to implicate apoptosis induction in anticarcinogenic activity of OSCs, the sequence of events leading to proapoptosis effects of OSCs is poorly defined.

The present study was undertaken to gain insights into the mechanism of apoptosis induction by OSCs (DAS, DADS and DATS) using PC-3 and DU145 human prostate cancer cells as a model. Prostate cancer cells were selected as a model because despite compelling epidemiological evidence for an inverse correlation between dietary intake of Allium vegetables and the risk for prostate cancer (Hsing et al., 2002) activity of OSCs against prostate cancer has not been examined. Here, we demonstrate that DATS is a significantly more potent inhibitor of PC-3 and DU145 cell proliferation and apoptosis inducer than DAS or DADS. We also provide evidence for important roles of Bcl-2, c-Jun N-terminal kinases (JNK1/2) and extracellular-signal regulated kinases (ERK1/2) in DATS-induced apoptosis.

Results

DATS inhibits proliferation of PC-3 cells by causing apoptosis

The effects of the OSC analogues on PC-3 cell viability were determined by sulforhodamine B assay, and the results are shown in Figure 1a. The viability of PC-3 cells was reduced significantly upon a 24-h exposure to DATS in a concentration-dependent manner with an IC50 of about 22 μ M (Figure 1a). Survival of PC-3 cells was also reduced in the presence of DADS (IC50 ≈35 μ M), whereas DAS was minimally active. For example, about 97% of the PC-3 cells were viable following a 24-h exposure to 40 μ M DAS, whereas only about 23% of the cells survived under similar conditions of DATS treatment (Figure 1a). It is interesting to note that the viability of PC-3 cells was not affected in the presence of dipropyl sulfide (DPS) or dipropyl disulfide (DPDS) even at 160 μ M concentration (Figure 1b). Collectively, these results indicated that the presence of allyl groups and the oligosulfide chain length affect activity of OSCs against proliferation of PC-3 cells. The growth inhibitory effects of the OSC analogues were confirmed by trypan blue dye exclusion assay (Figure 1c). In agreement with the results of sulforhodamine B assay, proliferation of PC-3 cells was suppressed in the presence of DATS in a concentration- and time-dependent manner as evidenced by a statistically significant decrease in the number of viable cells at 20 and 40 μ M concentrations at each time point compared with DMSO treated control (Figure 1c, bottom panel). Proliferation of PC-3 cells was minimally affected by DAS (Figure 1c, top panel), and an intermediate response was observed for DADS (Figure 1c, middle panel).

Figure 1
figure1

Sulforhodamine B assay for the effects of (a) DAS, DADS and DATS, and (b) DPS and DPDS on PC-3 cells viability. PC-3 cells were treated with DMSO (control) or different concentrations of the desired OSC analogue for 24 h. (c) Trypan blue dye exclusion assay for the effects of DAS, DADS and DATS on proliferation of PC-3 cells. PC-3 cells were treated with DMSO (control) or different concentrations of the desired OSC analogue for 24, 48 or 72 h. Data are mean±s.e. (n=3). *Significantly different compared with control (P<0.05) by one-way ANOVA

Next, we addressed the question whether growth-suppressive effects of the OSC analogues correlated with apoptosis induction, which is a widely accepted mechanism for antiproliferative activity of many naturally occurring as well as synthetic agents (Ferreira et al., 2002). Apoptosis (also known as programmed cell death) is a genetically regulated process of cellular suicide that is characterized by a number of morphological and cellular changes, including chromatin condensation, membrane blebbing, DNA fragmentation and cleavage of key cellular proteins (Hengartner, 2000). Apoptosis-inducing effects of OSC analogues were assessed by terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL) and by fluorescence microscopy following staining with 4′,6-diamidino-2-phenylindole (DAPI). As shown in Figures 2a and b, a 24-h treatment of PC-3 cells with 40 μ M DATS resulted in a statistically significant increase in the percentage of brown-color TUNEL-positive apoptotic bodies when compared with the control. In comparison with DMSO-treated control, the percentage of TUNEL-positive cells was increased by about nine-fold in DATS-treated cultures (Figure 2b). Similar treatment of PC-3 cells with DAS and DADS (40 μ M for 24 h) resulted in about 2.6- and 4.3-fold increase, respectively, in the percentage of brown-color apoptotic bodies when compared with control (Figure 2b). The fraction of cells with condensed and fragmented nuclei (DAPI assay) was about 13-fold higher in PC-3 cultures treated with 40 μ M DATS for 24 h than in the DMSO-treated controls (Figures 2a and b). Similar treatment of PC-3 cells with DAS and DADS (40 μ M for 24 h) caused an approximate 1.5- and 3.3-fold increase in percentage of cells with fragmented nuclei, respectively, compared with control (Figure 2b).

Figure 2
figure2

(a) Analysis of apoptotic cells by TUNEL and DAPI assay in PC-3 cultures treated with DMSO (control) and DATS (40 μ M) for 24 h. The figures show a marked increase in brown-color apoptotic bodies (TUNEL-positive) and cells with condensed and segmented nuclei (DAPI) in DATS treated PC-3 cultures compared with control. (b) Quantitation of apoptotic cells in PC-3 cultures exposed for 24 h to DMSO (control) or 40 μ M DAS, DADS or DATS. Data are mean±s.e. (n=3 or more). (c) Sub-G0/G1 fraction following a 24 h exposure of PC-3 cells to DMSO (control) or different concentrations of DAS, DADS or DATS. Data are mean±s.e. of three determinations. *Significantly different compared with control (P<0.05) by one-way ANOVA

Flow cytometric analysis of cells with sub-G0/G1 DNA content following staining with propidium iodide is a widely accepted technique for apoptosis detection. As can be seen in Figure 2c, a 24-h treatment of PC-3 cells with DATS resulted in a statistically significant increase in the fraction of sub-G0/G1 cells compared with control at both 20 and 40 μ M concentrations. For instance, in DMSO-treated control roughly 4% of the cells exhibited sub-G0/G1 DNA content, which was increased by >3-fold upon a 24-h exposure to 40 μ M DATS (Figure 2c). These results indicated that antiproliferative activity of OSCs against PC-3 cells correlated with their ability to induce apoptotic cell death.

DATS treatment causes Bcl-2 phosphorylation in PC-3 cells

Members of the Bcl-2 protein family play critical roles in regulation of apoptosis by acting as promoters (e.g. Bax) or inhibitors (e.g. Bcl-2) of the cell death process (Adams and Cory, 1998). Immunoblotting using antibodies against Bcl-2, BID and Bax was performed to gain insights into the mechanism of DATS-induced apoptosis. As can be seen in Figure 3a, a decrease of about 34–52% in the level of Bcl-2 protein was observed in DATS-treated cells at 16 and 24 h time points compared with control, which was accompanied by emergence of an immunoreactive band with reduced electrophoretic mobility (Figure 3a). On the other hand, DATS treatment did not alter the protein level of BID or Bax (Figure 3a).

Figure 3
figure3

(a) Immunoblotting for Bcl-2, BID and Bax using lysates from PC-3 cells cultured in the presence of 40 μ M DATS for the indicated times. Equal amounts of lysate protein from control and DATS-treated cells were used for gel electrophoresis. Blot was stripped and reprobed with antibody against actin to ensure equal protein loading. (b) Immunoblotting for Bcl-2 using lysates from PC-3 cells treated for 24 h with DMSO (control) or 40 μ M DATS with or without pre-treatment with λ-protein phosphatase. Blot was stripped and reprobed with antibody against actin to ensure equal protein loading. (c) Immunoblotting for Thr56 phosphorylated Bcl-2 using Bcl-2 immunoprecipitates from control and DATS treated (40 μ M for 4 or 24 h) cell lysates. (d) Effect of DATS treatment on Bcl-2 : Bax interaction. Aliquots containing 200 μg protein from lysates of control and DATS-treated cells (40 μ M for 4 or 24 h) were subjected to immunoprecipitation using antibody against Bcl-2 followed by immunoblotting using anti-Bax antibody. (e) Immunoblotting for cleaved caspase-3 and procaspase-9 using lysates from PC-3 cells exposed to 40 μ M DATS for the indicated times. The experiments were repeated two or three times, and the results were comparable

Electrophoretic mobility retardation of a protein is often indicative of post-translational modification (e.g. phosphorylation). To determine if the immunoreactive band with reduced electrophoretic mobility was a phosphorylated form of Bcl-2, the lysates from control and DATS-treated cells (40 μ M for 24 h) were treated with λ-protein phosphatase (a vanadate/EDTA-inhibitable phosphatase that can dephosphorylate phosphoserine, phosphothreonine and phosphotyrosine) prior to immunoblotting for Bcl-2. As can be seen in Figure 3b, the Bcl-2 immunoreactive band with reduced electrophoretic mobility was eliminated upon treatment of DATS-treated lysate with λ-protein phosphatase indicating phosphorylation of Bcl-2 in DATS-treated cells. Phosphorylation of Bcl-2 in DATS-treated cells was confirmed by immunoprecipitation using anti-Bcl-2 antibody followed by immunoblotting using anti-phospho-Bcl-2 (Thr56) antibody. As can be seen in Figure 3c, treatment of PC-3 cells with 40 μ M DATS for 24 h resulted in an approximate 5.4-fold increase in the level of Thr56 phosphorylated Bcl-2 when compared with control (Figure 3c).

The effect of phosphorylation of Bcl-2 on its antiapoptotic function remains ambiguous, but some studies have suggested that phosphorylated Bcl-2 loses its ability to heterodimerize with Bax (Haldar et al., 1996). To determine if DATS treatment affected the Bcl-2 : Bax interaction, Bcl-2 immunoprecipitate was subjected to immunoblotting using anti-Bax antibodies. As shown in Figure 3d, a noticeable decrease in Bax protein level was observed in the Bcl-2 immunoprecipitate from DATS-treated lysates compared with control. These results indicated that DATS treatment reduced heterodimer complex formation between Bcl-2 and Bax.

Next, we explored the possibility whether reduced Bcl-2 : Bax interaction resulted in activation of mitochondrial caspase cascade. As can be seen in Figure 3e, treatment of cells with 40 μ M DATS resulted in cleavage of procaspase-3 as evidenced by appearance of 19 and 17 kDa cleavage intermediates at 16 and 24 h time points. Caspase-3 is an executioner caspase that can be activated by a mitochondrial pathway involving caspase-9 (Thornberry and Lazebnik, 1998). Level of procaspase-9 was reduced by DATS treatment, which preceded caspase-3 activation (Figure 3e). These results indicated that DATS treatment triggers the mitochondrial mediated caspase cascade.

The effects of DAS and DADS on protein levels of Bcl-2, BID and Bax were also determined. DAS or DADS treatment did not alter protein level of Bcl-2, BID or Bax (data not shown). Since DATS was a more potent apoptosis inducer than other OSCs, this agent was selected for further investigation of the mechanism of cell death.

Role of mitogen-activated protein kinases in DATS-induced Bcl-2 phosphorylation

Since previous studies have implicated mitogen-activated protein kinases (MAPKs) in phosphorylation of Bcl-2 in some systems (Srivastava et al., 1999; Yamamoto et al., 1999; Blagosklonny, 2001), the question was raised whether MAPKs contributed to DATS-induced Bcl-2 phosphorylation in our model. We addressed this question by examining the effect of DATS on activation ERK1/2, JNK and p38 MAPK. As shown in Figure 4a, exposure of PC-3 cells to 40 μ M DATS resulted in a rapid but transient activation (phosphorylation) of ERK1/2. Activation of ERK1/2 was evident as early as 15 min after treatment, reached a maximum between 15 and 30 min, and declined thereafter (Figure 4a, top panel). Protein level of ERK1/2 was not affected by DATS treatment (Figure 4b, top panel). The effect of DATS on kinase activity of ERK1/2 was determined by examining Ser383 phosphorylation of Elk-1, which is one of the downstream targets of ERK1/2. In comparison with DMSO-treated control cells, Ser383 phosphorylation of Elk-1 was increased by about 15-fold upon a 30 min treatment with 40 μ M DATS (Figure 4c, top panel). DATS treatment did not cause a change in the level of phosphorylated or total p38 MAPK (Figures 4a and b, middle panel). Interestingly, treatment of PC-3 cells with DATS also resulted in rapid but sustained activation of JNK1. Activation of JNK1 was observed as early as 30 min after treatment and persisted for the duration of the experiment (Figure 4a, bottom panel). Similar to ERK1/2 and p38, however, DATS treatment did not alter the level of JNK1 protein (Figure 4b, bottom panel). The effect of DATS treatment on kinase activity of JNK was determined by examining Ser63 phosphorylation of c-Jun, which is a downstream target of JNKs. An approximate two-fold increase in Ser63 phosphorylation of c-Jun was observed in cells treated with 40 μ M DATS for 4 h (Figure 4c, bottom panel). Together, these observations indicated that DATS treatment caused activation of ERK1/2 and JNK1, but not p38 MAPK.

Figure 4
figure4

Immunoblotting for (a) phosphorylated (activated), and (b) total ERK1/2 (top panel), p38 (middle panel) and JNK1 (bottom panel) using lysates from PC-3 cells treated with 40 μ M DATS for the indicated times. Equal amounts of lysate proteins were used for gel electrophoresis. Blots were stripped and reprobed with anti-actin antibody to ensure equal protein loading. (c) ERK1/2 and JNK kinase activities using lysates from PC-3 cells treated with DMSO (control) and 40 μ M DATS for 30 min (ERK1/2) or 4 h (JNK). The experiment was repeated two or three times to ensure reproducibility

To experimentally verify the roles of ERK1/2 and JNK1 in DATS-induced Bcl-2 phosphorylation and apoptosis, studies were carried out using inhibitors of MEK1 (an upstream kinase in ERK1/2 signaling pathway), p38 MAPK and JNK. DATS-induced phosphorylation of Bcl-2 was abolished in the presence of JNK specific inhibitor SP600125, as evidenced by disappearance of the Bcl-2 immunoreactive band with reduced electrophoretic mobility with a concomitant increase in the level of the faster migrating unphosphorylated form of Bcl-2 (Figure 5a). MEK1 inhibitor PD98059 afforded partial protection against DATS-induced Bcl-2 phosphorylation (Figure 5a). DATS-induced phosphorylation of Bcl-2 was not affected in the presence of p38 MAPK-specific inhibitor SB202190 (data not shown).

Figure 5
figure5

Effects of PD98059 (MEK1 inhibitor) and SP600125 (JNK inhibitor) on DATS-induced Bcl-2 phosphorylation and apoptosis. PC-3 cells were pre-treated with the desired MAPK inhibitor (50 μ M PD98059 or 20 μ M SP600125) for 2 h, and then exposed to 40 μ M DATS for 24 h in the presence of the inhibitor. Cells were harvested and processed for (a) immunoblotting for Bcl-2, (b) DAPI assay, and (c) analysis of sub-G0/G1 cells. Data are mean±s.e. of three determinations. Statistical significance was assessed by one-way ANOVA followed by Bonferroni's multiple comparison test. *Significantly different compared with control (P<0.05). +Significantly different compared with DATS alone group (P<0.05)

Next, we determined the effects of MAPK inhibitors on DATS-induced apoptosis by DAPI assay, and the results are shown in Figure 5b. Apoptosis induction by DATS was significantly attenuated upon treatment of cells with JNK-specific inhibitor SP600125. For instance, in DATS-treated PC-3 cultures (40 μ M for 24 h) roughly 35% of the cells exhibited nuclear fragmentation, which was reduced to only about 3% in the presence of SP600125 (Figure 5b). The percentage of cells with fragmented nuclei was also reduced significantly in the presence of MEK1 inhibitor PD98059, although the protection was much less pronounced when compared with JNK inhibitor. The p38 MAPK inhibitor SB202190 did not protect against DATS-induced apoptosis (data not shown). The effects of MAPK inhibitors on DATS-induced apoptosis were further examined by quantifying cells with sub-G0/G1 DNA content (Figure 5c). In agreement with the results of DAPI assay, DATS-induced accumulation of sub-G0/G1 cells was reduced significantly in the presence of both SP600125 and PD98059. However, a greater protection was observed in the presence of JNK inhibitor than in the presence of MEK1 inhibitor (Figure 5c). These results indicated that activation of JNK1, and to some extent ERK1/2, contributed to DATS-induced Bcl-2 phosphorylation and apoptosis in PC-3 cells.

Effect of Bcl-2 overexpression on DATS-induced apoptosis

To further investigate the role of Bcl-2 in DATS-induced apoptosis, sensitivity of PC-3 cells stably transfected with Bcl-2 (PC-3/Bcl-2) and vector-transfected control cells (PC-3/neo) to apoptosis induction by DATS was compared by DAPI assay, and the results are shown in Figure 6. As can be seen in Figure 6a, the level of Bcl-2 protein was significantly higher in PC-3/Bcl-2 cells than in PC-3/neo. Similar to untransfected PC-3 cells (Figure 2a), a 24-h exposure of PC-3/neo cells to DATS resulted in significant increase in percentage of apoptotic cells when compared with DMSO-treated control (Figures 6b and c). On the other hand, PC-3/Bcl-2 cells were significantly more resistant to DATS-induced apoptosis compared with PC-3/neo cells at both 20 and 40 μ M concentrations (Figures 6b and c). Analysis of cytoplasmic histone-associated DNA fragments confirmed that Bcl-2 overexpression in PC-3 cells conferred protection against DATS-induced cell death (data not shown). These results underscore the importance of Bcl-2 in cell death induced by DATS.

Figure 6
figure6

(a) Immunoblotting for Bcl-2 using lysates from untreated PC-3/neo and PC-3/Bcl-2 cells. Blot was stripped and reprobed with anti-actin antibody to ensure equal protein loading. (b) Analysis of apoptosis by DAPI assay in PC-3/neo and PC-3/Bcl-2 cells cultured in the presence of DMSO (control) or 40 μ M DATS for 24 h. (c) Quantitation of apoptotic cells by DAPI assay following a 24 h exposure of PC-3/neo or PC-3/Bcl-2 cells to DMSO (control) and 20 or 40 μ M DATS. Data are mean±s.e. (n=3). *Significantly different (P<0.05) compared with respective DMSO-treated control, and +Significantly different (P<0.05) compared with DATS-treated PC-3/neo. Statistical significance was assessed by one-way ANOVA followed by Bonferroni's multiple comparison test

DATS causes apoptosis and activates ERK1/2 and JNK1/2 in DU145 cells

To determine whether the effects of DATS described above were unique to the PC-3 cell line, effect of DATS on cell viability, apoptosis induction and activation of ERK1/2 and JNKs was examined in human prostate adenocarcinoma DU145 cells. Similar to PC-3 cells, proliferation of DU145 cells was inhibited significantly in the presence of 40 μ M DATS, whereas DAS was practically inactive and an intermediate response was observed for DADS (Figure 7a). However, DU145 cells were relatively more sensitive to DADS in comparison with PC-3 cells (compare Figures 1c and 7a). Similar to PC-3 cells, DU145 cell line was significantly more sensitive to cell death by DATS compared with DAS or DADS as determined by DAPI assay (Figure 7b) as well as analysis of sub-G0/G1 cells (data not shown). DATS treatment caused rapid but transient activation of JNK1/2 in DU145 cells (Figure 7c, top panel). Exposure of DU145 cells to 40 μ M DATS also resulted in rapid but transient activation of ERK1/2 (Figure 7c, middle panel). Activation of ERK1/2 and JNK1/2 in DATS-treated DU145 cell line was respectively coupled to increased phosphorylation of Elk-1 (Ser383) and c-jun (Ser63) (Figure 7d). Similar to the results in PC-3 cells, DATS treatment did not alter protein level or phosphorylation of p38 MAPK in DU145 cells (data not shown). In addition, immunoblotting for Bcl-2 using lysates from DATS-treated DU145 cells revealed a band with reduced electrophoretic mobility at 8 and 16 h time points (data not shown). These results indicated that DATS-induced Bcl-2 phosphorylation and activation of ERK1/2 and JNK1/2 are not unique to PC-3 cells.

Figure 7
figure7

(a) Effects of OSCs on proliferation of DU145 cells determined by trypan blue dye exclusion assay. DU145 cells were treated with DMSO (control) or 40 μ M DAS, DADS or DATS for 24, 48 or 72 h. Data are mean±s.e. (n=3). *Significantly different compared with control (P<0.05) by one-way ANOVA. (b) Effects of OSCs on apoptosis induction in DU145 cells determined by DAPI assay. DU145 cells were treated with DMSO (control) or 40 μ M DAS, DADS or DATS for 24 h. Data are mean±s.e. (n=3). *Significantly different compared with control (P<0.05) by one-way ANOVA. (c) Immunoblotting for phospho-JNK1/2 (top panel) and phospho-ERK1/2 (middle panel) using lysates from DU145 cells treated with 40 μ M DATS for the indicated times. Blots were stripped and reprobed with anti-actin antibody (bottom panel) to ensure equal protein loading. Results were similar in three independent experiments, and representative data from a single experiment are shown. (d) Effect of DATS treatment (40 μ M, 30 min) on kinase activity of ERK1/2 and JNKs determined by monitoring phosphorylation of Elk-1 (Ser383) and c-Jun (Ser63), respectively. The experiment was repeated two times, and the results were comparable. (e) Effect of MEK1 inhibitor U0126 on apoptosis induction by DATS in DU145 cells determined by quantitation of cytoplasmic histone-associated DNA fragments. DU145 cells were pretreated with U0126 (10 μ M) for 2 h, and then treated with 40 μ M DATS for 24 h in the presence of the inhibitor. Cells were collected, and processed for analysis of cytoplasmic histone-associated DNA fragments. Experiment was repeated three times, and the results were similar. Representative data (mean±s.e.) from a single experiment are shown. *Significantly different compared with DMSO-treated control, P<0.05. (f) Effect of JNK inhibitor SP600125 on apoptosis induction by DATS in DU145 cells. DU145 cells were pretreated with SP600125 (10 μ M) for 2 h, and then treated with 40 μ M DATS for 24 h in the presence of the inhibitor. Cells were collected, and processed for analysis of cytoplasmic histone-associated DNA fragments. Experiment was repeated three times, and the results were similar. Representative data (mean±s.e.) from a single experiment are shown. *Significantly different compared with DMSO-treated control, P<0.05, and +Significantly different compared with DATS alone group, P<0.05. Statistical significance was assessed by one-way ANOVA followed by Bonferroni's multiple comparison test

To further verify the roles of MAPKs in apoptosis induction by DATS, effects of pharmacological inhibition of ERK1/2 and JNK1/2 using U0126 and SP600125, respectively, on DATS-induced apoptosis were determined in DU145 cells by analysis of cytoplasmic histone-associated DNA fragments. As can be seen in Figure 7e, apoptosis induction by DATS was lower in cells treated with U0126 in comparison with cells treated with DATS alone. DATS-induced cytoplasmic histone-associated DNA fragmentation was reduced significantly by pharmacological inhibition of JNK (Figure 7f). These results confirmed that activation of JNK1/2, and to some extent ERK1/2, contributed to DATS-induced apoptosis in DU145 cells as well.

Catalase overexpression inhibits JNK activation and apoptosis induction by DATS

Previous studies have shown that JNK-mediated apoptosis in some systems is associated with generation of hydrogen peroxide as a second messenger (Saitoh et al., 1998; Wang et al., 1998). We raised the question whether DATS-induced JNK activation involves hydrogen peroxide as a second messenger. To examine this possibility systematically, we infected DU145 cells with adenoviral vectors containing catalase (Ad-catalase) or enhanced green fluorescence protein (Ad-EGFP), and compared their sensitivity to DATS-mediated activation of JNK1/2 and ERK1/2, and apoptosis induction. As shown in Figure 8a, catalase protein expression was significantly higher in DU145 cells infected with Ad-catalase than in the cells infected with Ad-EGFP. An average infection efficiency of more than 70% was achieved in different experiments. DATS treatment (40 μ M, 30 min) did not alter the protein level of either endogenous or exogenously expressed catalase (Figure 8a). Similar to parental DU145 cells, activation of both JNK1/2 and ERK1/2 was evident in DATS treated (40 μ M, 30 min) Ad-EGFP cells. While catalase overexpression inhibited DATS-mediated JNK1/2 activation, ERK1/2 activation was not significantly affected (Figure 8a). Protein level of JNK1/2 or ERK1/2 was not altered by DATS treatment in cells infected with either Ad-EGFP or Ad-catalase (data not shown). As can be seen in Figure 8b, catalase overexpression significantly inhibited DATS-induced apoptosis. For example, cytoplasmic histone-associated DNA fragmentation in DATS-treated (40 μ M, 24 h) Ad-EGFP cells was higher by about 2.4-fold in comparison with DMSO-treated control, whereas an increase of only about 30% in DNA fragmentation over control was observed in catalase overexpressing DU145 cells. These results clearly indicated that DATS-mediated activation of JNK1/2, but not ERK1/2, involves generation of hydrogen peroxide as a second messenger.

Figure 8
figure8

(a) Immunoblotting for catalase, phospho-JNK1/2, and phospho-ERK1/2 using lysates from Ad-EGFP or Ad-catalase DU145 cells treated with DMSO (control) or 40 μ M DATS for 30 min. The blots were stripped and reprobed with anti-actin antibodies to ensure equal protein loading. Experiment was repeated at least three times, and the results were comparable. Data from a representative experiment are shown. (b) Analysis of cytoplasmic histone-associated DNA fragments following a 24 h treatment of Ad-EGFP or Ad-catalase DU145 cells with DMSO (control) or 40 μ M DATS. Experiment was repeated three times with similar results, and representative data from a single experiment are shown. Data are mean±s.e. (n=3). *Cytoplasmic histone-associated DNA fragmentation was significantly higher in DATS-treated cells compared with DMSO control in Ad-EGFP but not in Ad-catalase

Discussion

Data presented herein indicate that PC-3 and DU145 cells are highly sensitive to growth inhibition by DATS, and presence of allyl groups and the oligosulfide chain length affect activity of OSCs against proliferation of human prostate cancer cells. The growth-suppressive effects of OSCs against PC-3 and DU145 cells correlated with their ability to induce apoptosis. Apoptosis induction was significantly higher in cells treated with DATS compared with DAS or DADS indicating important role for the oligosulfide chain length in OSC-induced apoptosis.

The Bcl-2 family proteins play critical roles in regulation of cell death processes (Adams and Cory, 1998). In addition, phosphorylation of Bcl-2 has emerged as an important mechanism for regulation of its function (May et al., 1994; Haldar et al., 1995, 1996; Ito et al., 1997; Fang et al., 1998; Ruvolo et al., 1998; Srivastava et al., 1999). In our model, DATS treatment caused phosphorylation of Bcl-2. Functional significance of Bcl-2 phosphorylation in regulation of apoptosis remains ambiguous. Bcl-2 phosphorylation resulting from treatment with interleukin-3 or protein kinase C agonists has been shown to enhance antiapoptotic function of Bcl-2 (May et al., 1994; Ito et al., 1997; Ruvolo et al., 1998). On the other hand, an abrogation in antiapoptotic function of Bcl-2 due to its phosphorylation has been demonstrated in cells treated with Taxol or other microtubule-targeting anticancer agents (Haldar et al., 1995, 1996; Srivastava et al., 1999). Bcl-2 binds to and inactivates proapoptotic protein Bax, but whether or not Bcl-2 phosphorylation affects its interaction with Bax is controversial. In some systems, phosphorylated Bcl-2 loses its ability to form heterodimer with Bax (Haldar et al., 1996). On the other hand, Scatena et al. (1998) demonstrated that phosphorylated Bcl-2 remained complexed with Bax in cells undergoing apoptosis upon treatment with paclitaxel. In our model, DATS-induced Bcl-2 phosphorylation was associated with a marked decrease in heterodimer complex formation between Bcl-2 and Bax. Thus, it is reasonable to postulate that DATS-induced phosphorylation of Bcl-2 reduces its interaction with Bax to trigger mitochondrial caspase cascade. Consistent with this postulation, cleavage of procaspase-3 and -9 was observed in DATS-treated PC-3 cells. Relative resistance of Bcl-2 overexpressing PC-3 cells to cell death by DATS indicates further that Bcl-2 regulates DATS-induced apoptosis.

The signal transduction pathways mediating Bcl-2 phosphorylation has been the subject of intense research. Several kinases including Raf-1, protein kinase Cα, cAMP-dependent protein kinase, MAPKs and cyclin-dependent kinase have been implicated in phosphorylation of Bcl-2 (Blagosklonny et al., 1996; Maundrell et al., 1997; Srivastava et al., 1998; Ruvolo et al., 1998; Yamamoto et al., 1999; Rosini et al., 2000; Deng et al., 2001; Torcia et al., 2001; Vantieghem et al., 2002). DATS treatment resulted in activation of ERK1/2 and JNKs, but not p38 MAPK, in both cell lines. Activation of ERK1/2 was rapid but transient in both cell lines. Even though JNK1 activation by DATS treatment was rapid in both cell lines, activation was transient in DU145 cells but sustained in PC-3. Mechanism of differences in DATS-mediated JNK activation kinetics between PC-3 and DU145 cells remains to be elucidated. The DATS-mediated activation of ERK1/2 and JNK1 in both cells lines was coupled to increased phosphorylation of their downstream targets Elk-1 and c-jun, respectively. Studies with inhibitors of MAPKs indicated that DATS-induced phosphorylation of Bcl-2 was mediated largely by JNK1 and to some extent by ERK1/2. DATS-induced Bcl-2 phosphorylation in PC-3 cells was fully abolished in the presence of JNK-specific inhibitor SP600125, which was evidenced by disappearance of slower migrating Bcl-2 immunoreactive band. Moreover, pharmacological inhibition of JNK afforded significant protection against DATS-induced apoptosis in both PC-3 and DU145 cells. The DATS-induced apoptosis and/or Bcl-2 phosphorylation were moderately inhibited by pharmacological inhibition of ERK1/2.

In this study, we observed that overexpression of an H2O2 scavenger, catalase, inhibited DATS-induced JNK activation as well as apoptotic death. These results suggest that DATS-induced JNK activation is mediated through oxidative stress. These observations are consistent with results obtained with DADS, which may represent an oxidizing agent able to induce oxidative stress-mediated cell death (Filomeni et al., 2003). A fundamental question, which remains unanswered, is how DATS induces oxidative stress and triggers signal transduction pathways. Previous studies illustrate that oxidative stress can be generated through a serious imbalance between the production of reactive oxygen species (ROS) and antioxidant defenses (Halliwell, 1996). Studies have shown that DADS treatment does not alter glutathione redox state or related enzymes activity (Sheen et al., 1996). These results indicate that generation of ROS rather than alteration of antioxidant defense system may be responsible for DATS-induced oxidative stress. The increase in steady-state levels of intracellular ROS can be detected through redox-sensing molecules including thioredoxin (TRX) and glutaredoxin (GRX). These molecules bind to apoptosis signal-regulating kinase 1 (ASK1) and suppress its activation (Saitoh et al., 1998; Song et al., 2002). TRX and GRX dissociate from ASK1 during oxidative stress and their dissociation from ASK1 activates the ASK1-SEK1-JNK1 signal transduction pathway (Saitoh et al., 1998; Song et al., 2002). Interestingly, our recent studies suggest that oxidative stress activates ASK1 through two different pathways: the glutathione-dependent GRX-ASK1 pathway and the glutathione-independent TRX-ASK1 pathway (Song and Lee, 2003). We postulate that DATS treatment increases the intracellular level of ROS, which can be detected by redox-sensing molecules, in particular TRX. We believe that this model will provide a framework for future studies.

Activation of JNK by DADS in a neuroblastoma cell line has been demonstrated recently (Filomeni et al., 2003). We did not observe activation of JNK1 or other MAPKs in PC-3 cells treated with 40 μ M DADS or DAS (D Xiao and SV Singh, unpublished observations). While the reasons for this discrepancy are not clear, DADS-induced activation of JNK1 in SH-SY5Y neuroblastoma cells could be a cell line-specific effect. Alternatively, DADS concentrations higher than 40 μ M might be required for activation of JNK1 in PC-3 cells. In another study, Xiao et al. (2003a) showed that treatment of SW480 human colon cancer cells with a rather high concentration (300 μ M) of S-allyl mercaptocysteine, a water-soluble OSC analogue, resulted in rapid and sustained activation of JNK1. Interestingly, pharmacological inhibition of JNKs using SP600125 attenuated the early phase (24 h) but not late phase (48 h) of apoptosis induced by S-allyl mercaptocysteine suggesting involvement of additional mechanisms in apoptosis induction by this analogue. Activation of ERK1/2 and p38 MAPK was also observed in S-allyl mercaptocysteine-treated SW480 cells (Xiao et al., 2003a). Unlike the results of the present study, however, pharmacological inhibition of ERK1/2 did not offer any protection against S-allyl mercaptocysteine-induced apoptosis in SW480 cells (Xiao et al., 2003a). Even though the two studies cited above indicated JNK1 activation in apoptosis induction by OSC analogues, it was not clear how JNK activation contributed to OSC-mediated apoptosis. Our data indicate that, at least in prostate cancer cells, DATS-mediated activation of JNK1 leads to Bcl-2 phosphorylation to trigger the mitochondria-mediated caspase cascade. To the best of our knowledge, the present study is the first published report to implicate Bcl-2 phosphorylation in the apoptosis-inducing activity of garlic-derived OSCs.

The antiproliferative and apoptosis-inducing effects of DATS against PC-3 and DU145 cells were observed at micromolar concentrations. It is difficult to predict if micromolar concentrations of DATS are achievable in humans since no pharmacokinetics data are available for this OSC analogue. However, the pharmacokinetics of S-allyl cysteine, a water-soluble OSC analogue, has been investigated in rodents and dogs (Nagae et al., 1994). This study showed that (a) S-allyl cysteine was absorbed rapidly, and reached peak plasma concentration within 0.5–1 h, and (b) the plasma concentrations following oral administration of 12.5, 25 and 50 mg S-allyl cysteine/kg body weight were about 50, 112 and 227 μ M, respectively. Further studies are needed to determine the bioavailability of DATS, and to determine if DATS is effective in inhibiting growth of human prostate cancer cells in in vivo models.

In conclusion, the present study demonstrates that (a) human prostate cancer cells are highly sensitive to growth inhibition and apoptosis induction by DATS, and that even subtle change in OSC structure (e.g. the oligosulfide chain length) could have a significant impact on its antiproliferative activity, (b) DATS-induced apoptosis is associated with phosphorylation of Bcl-2, which is mediated largely by JNK1 and to some extent by ERK1/2, (c) Bcl-2 overexpression in PC-3 cells confers significant protection against cell death induced by DATS, (d) relationship between apoptosis induction and ERK1/2 and JNK1 activation by DATS is not unique to PC-3 cells, and (e) DATS-induced JNK activation may involve hydrogen peroxide as a second messenger. However, the precise mechanism of JNK activation in DATS-treated prostate cancer cells remains to be elucidated.

Materials and methods

Reagents

DATS was purchased from LKT Laboratories (St. Paul, MN, USA). Other OSC analogues, including DAS, DADS, DPS and DPDS were from Aldrich (Milwaukee, WI, USA). Tissue culture media and fetal bovine serum were from GIBCO (Grand Island, NY, USA), propidium iodide and DAPI were from Sigma (St Louis, MO, USA), RNaseA was from Promega (Madison, WI, USA), and λ-protein phosphatase was from New England Biolabs (Beverly, MA, USA). ApopTag Peroxidase In situ Apoptosis Detection kit was purchased from Chemicon International (Temecula, CA, USA). Cytoplasmic histone-associated DNA fragmentation was measured using an ELISA kit from Roche Diagnostics (Mannheim, Germany). Antibodies against Bax (sc-493), BID (sc-6538), Thr56 phosphorylated Bcl-2 (sc-16321), ERK1 (sc-93), phospho-(Tyr204)-ERK1/2 (sc-7383), JNK1 (sc-571), phospho-(Thr183/Tyr185)-JNK (sc-12882), p38 (sc-535) and phospho-(Tyr182)-p38 (sc-7973) were from Santa Cruz Biotechnology (Santa Cruz, CA, USA), antibody against Bcl-2 was from DAKO (Carpinteria, CA, USA), antibody specific for cleaved caspase-3 (catalogue # 9661) was from Cell Signaling Technology (Beverly, MA, USA), antibody against procaspase-9 (catalogue # 556585) was from BD Pharmingen, Palo Alto, CA, USA and the antibody against actin was from Oncogene Research Products (Boston, MA, USA). MEK1 inhibitor 2′-amino-3′-methoxyflavone (PD98059), p38 MAPK inhibitor 4-(4-fluorophenyl)-2-(4-hydroxyphenyl)-5-(4-pyridyl)1H-imidazole (SB202190), and JNK inhibitor anthra[1,9-cd]pyrazol-6(2 H)-one-1,9-pyrazoloanthrone (SP600125) were purchased from Calbiochem (La Jolla, CA, USA). The assay kits for p44/42 (ERK1/2) and JNK kinases were from Cell Signaling Technology (Beverly, MA, USA).

Cell culture and cell survival assay

Monolayer cultures of PC-3 cells were maintained in F-12 K Nutrient Mixture (Kaighn's Modification) supplemented with 7% (v/v) non-heat inactivated fetal bovine serum and antibiotics. The PC-3/neo and PC-3/Bcl-2 cells, a generous gift from Dr Natasha Kyprianou (University of Maryland School of Medicine, Baltimore, MD, USA), were maintained similarly except that 500 μg/ml G418 was added to the cultures. DU145 cultures were maintained in Eagle's minimum essential medium supplemented with 1 mM sodium pyruvate, 0.1 mM nonessential amino acids, 1.5 g/l sodium bicarbonate, and 10% (v/v) fetal bovine serum and antibiotics. Each cell line was maintained in a humidified atmosphere of 95% air and 5% CO2 at 37°C. Effect of OSC analogues on cell proliferation was determined by sulforhodamine B and trypan blue dye exclusion assay. Sulforhodamine B assay was performed as described by us previously (Xiao and Singh, 2002). For trypan blue dye exclusion assay, 5 × 103 cells in 1 ml complete medium were plated in six-well plates, and allowed to attach overnight. The medium was replaced with fresh complete medium containing different concentrations of the desired OSC analogue, and the plates were incubated for 24, 48 or 72 h at 37°C. Stock solutions of the OSCs were prepared in DMSO, and an equal volume of DMSO (final concentration 0.1%) was added to the control wells. At the end of the incubation, both floating and adherent cells were collected and suspended in 25 μl of phosphate-buffered saline (PBS). The cells were then mixed with 5 μl of 0.4% trypan blue solution, and live (unstained) and dead (stained) cells were counted under an inverted microscope.

Assessment of apoptosis

Apoptosis induction in OSC-treated cells was assessed by (a) TUNEL assay, (b) fluorescence microscopy analysis of cells with condensed and segmented nuclei following staining with DAPI, (c) flow cytometric analysis of cells with sub-G0/G1 DNA content following staining with propidium iodide, or (d) quantitation of cytoplasmic histone-associated DNA fragments. TUNEL assay was performed using ApopTag Peroxidase In situ Apoptosis Detection kit (Chemicon International, Temecula, CA, USA) according to the manufacturer's instructions with slight modifications. Briefly, 2 × 104 cells were grown on coverslips, and allowed to attach overnight. Cells were then exposed to DMSO (control) or desired concentration of the OSC for 24 h, and fixed with 4% paraformaldehyde in PBS for 20 min at room temperature. After washing three times with PBS, cells were permeabilized with 0.2% Triton X-100 for 15 min. Cells were then processed for TUNEL assay according to the manufacturer's protocol, and brown color apoptotic bodies were visualized under a microscope at × 20 magnification. For DAPI staining, cells were treated and fixed with paraformaldehyde as described above for TUNEL assay. After rinsing with PBS, cells were stained with DAPI (10 μg/ml) for 5 min. Nuclear condensation and fragmentation was examined under a fluorescence microscope at × 20 magnification. In some experiments, cells were pretreated for 2 h with MEK1 inhibitor PD98059, p38 MAPK inhibitor SB202190, or JNK inhibitor SP600125, and then either left untreated (inhibitor alone group) or exposed to 40 μ M DATS for 24 h prior to processing for DAPI staining.

For analysis of cells with sub-G0/G1 DNA content, cells (5 × 105) were seeded in T25-flasks, and allowed to attach overnight. The medium was replaced with fresh complete medium containing desired concentrations of the OSCs. An equal volume of DMSO was added to the controls. After 24 h incubation at 37°C, floating and attached cells were collected, washed with PBS, and fixed with 70% ethanol. Fixed cells were then treated with 80 μg/ml of RNaseA and 50 μg/ml propidium iodide for 30 min, and stained cells were analysed using a Coulter Epics XL Flow Cytometer. In some experiments, cells were pretreated for 2 h with MAPK inhibitors prior to treatment with DATS followed by analysis of sub-G0/G1 cells. Cytoplasmic histone-associated DNA fragmentation was determined according to the manufacturer's instructions.

Immunoblotting and kinase assays

Cells were treated with the desired OSC or DMSO as described above, and lysed as reported by us previously (Xiao et al., 2003b). The cell lysate was cleared by centrifugation at 14 000 g for 15 min. Supernatant proteins were resolved by sodium-dodecyl sulfate-polyacrylamide gel electrophoresis (SDS–PAGE), and transferred onto PVDF membrane. After blocking with 10% nonfat dry milk in Tris-buffered saline containing 0.05% Tween-20, the membrane was incubated with the desired primary antibody for 1 h. The membrane was then treated with appropriate secondary antibody, and the immunoreactive bands were visualized using enhanced chemiluminescence kit (NEN Life Science Products, Boston, MA, USA). Each membrane was stripped and reprobed with antibody against actin to ensure equal protein loading. In some experiments, cells were pretreated for 2 h with MAPK inhibitors, and then either left untreated or exposed to 40 μ M DATS for 24 h prior to preparation of cell lysates and immunoblotting. The p44/42 (ERK1/2) and JNK kinase activities in the lysates from control and DATS-treated cells were determined using commercially available kits (Cell Signaling Technology, Beverly, MA, USA) according to the manufacturer's instructions.

λ-Protein phosphatase treatment

Aliquots containing 40 μg of lysate protein from cells exposed to 40 μ M DATS or DMSO (control) for 24 h were incubated with either λ-protein phosphatase (400 U, New England Biolabs, Beverly, MA, USA) or the control buffer at 30°C for 4 h in a phosphatase reaction buffer containing 2 mM MnCl2. Subsequently, the samples were subjected to immunoblotting for Bcl-2.

Immunoprecipitation

Cells were treated with DATS or DMSO as described above, washed twice with ice-cold PBS, and lysed. Aliquot containing 200 μg of lysate protein was incubated with 4 μg of anti-Bcl-2 antibody overnight at 4°C with gentle shaking. Protein A-agarose (50 μl, Santa Cruz Biotechnology, Santa Cruz, CA, USA) was then added to each sample, and the incubation was continued for an additional 3 h at 4°C. The immunoprecipitates were washed five times with lysis buffer, and subjected to SDS-PAGE followed by immunoblotting using antibody against Thr56 phosphorylated Bcl-2 or Bax.

Adenoviral infection

Adenoviral constructs containing EGFP and catalase have been described previously (Song et al.,. 2002). Cells (2.5 × 105) were plated in 60-mm plates, and infected after 24 h with Ad-EGFP or Ad-catalase (20 MOI). After 36 h incubation at 37°C, the cells were treated with DMSO or 40 μ M DATS for 30 min (for immunoblotting of catalase, total ERK1/2, total JNK1/2, phospho-JNK1/2 and phospho-ERK1/2) or for 24 h for analysis of cytoplasmic histone-associated DNA fragments. Cells were then collected and processed for immunoblotting or apoptosis assay.

References

  1. Adams JM and Cory S . (1998). Science (Washington, DC), 281, 1322–1326.

  2. Agarwal KC . (1996). Med. Res. Rev., 16, 111–124.

  3. Blagosklonny MV . (2001). Leukemia, 15, 869–874.

  4. Blagosklonny MV, Schulte T, Nguyen P, Trepel J and Neckers LM . (1996). Cancer Res., 56, 1851–1854.

  5. Dausch JG and Nixon DW . (1990). Prev. Med., 19, 346–361.

  6. Deng X, Xiao L, Lang W, Gao F, Ruvolo P and May WS . (2001). J. Biol. Chem., 276, 23681–23688.

  7. Dorant E, van den Brandt PA, Goldbohm RA, Hermus RJ and Sturmans F . (1993). Br. J. Cancer, 67, 424–429.

  8. Fang G, Chang BS, Kim CN, Perkins C, Thompson CB and Bhalla KN . (1998). Cancer Res., 58, 3202–3208.

  9. Ferreira CG, Epping M, Kruyt FAE and Giaccone G . (2002). Clin. Cancer Res., 8, 2024–2034.

  10. Filomeni G, Aquilano K, Rotilio G and Ciriolo MR . (2003). Cancer Res., 63, 5940–5949.

  11. Haldar S, Chintapalli J and Croce CM . (1996). Cancer Res., 56, 1253–1255.

  12. Haldar S, Jena N and Croce CM . (1995). Proc. Natl. Acad. Sci. USA, 92, 4507–4511.

  13. Halliwell B . (1996). Biochem. Soc. Trans., 24, 1023–1027.

  14. Hengartner MO . (2000). Nature (Lond.), 407, 770–776.

  15. Hsing AW, Chokkalingam AP, Gao YT, Madigan MP, Deng J, Gridley G and Fraumeni Jr JF . (2002). J. Natl. Cancer Inst., 94, 1648–1651.

  16. Ito T, Deng X, Carr B and May WS . (1997). J. Biol. Chem., 272, 11671–11673.

  17. Knowles LM and Milner JA . (1998). Nutr. Cancer, 30, 169–174.

  18. Knowles LM and Milner JA . (2001). J. Nutr., 131, 1061s–11066s.

  19. Kwon KB, Yoo SJ, Ryu DG, Yang JY, Rho HW, Kim JS, Park JW, Kim HR and Park BH . (2002). Biochem. Pharmacol., 63, 41–47.

  20. Maundrell K, Antonsson B, Magnenat E, Camps M, Muda M, Chabert C, Gillieron C, Boschert U, Vial-Knecht E, Martinou JC and Arkinstall S . (1997). J. Biol. Chem., 272, 25238–25242.

  21. May WS, Tyler PG, Ito T, Armstrong DK, Qatsha KA and Davidson NE . (1994). J. Biol. Chem., 269, 26865–26870.

  22. Nagae S, Ushijima M, Hatono S, Imai J, Kasuga S, Matsuura H, Itakura Y and Higashi Y . (1994). Planta Med., 60, 214–217.

  23. Nakagawa H, Tsuta K, Kiuchi K, Senzaki H, Tanaka K, Hioki K and Tsubura A . (2001). Carcinogenesis, 22, 891–897.

  24. Reddy BS, Rao CV, Rivenson A and Kelloff G . (1993). Cancer Res., 53, 3493–3498.

  25. Robert V, Mouille B, Mayeur C, Michaud M and Blachier F . (2001). Carcinogenesis, 22, 1155–1161.

  26. Rosini P, De Chiara G, Lucibello M, Garaci E, Cozzolino F and Torcia M . (2000). Biochem. Biophys. Res. Commun., 278, 753–759.

  27. Ruvolo PP, Deng X, Carr BK and May WS . (1998). J. Biol. Chem., 273, 25436–25442.

  28. Saitoh M, Nishitoh H, Fujii M, Takeda K, Tobiume K, Sawada Y, Kawabata M, Miyazono K and Ichijo H . (1998). EMBO J., 17, 2596–2606.

  29. Scatena CD, Stewart ZA, Mays D, Tang LJ, Keefer CJ, Leach SD and Pietenpol JA . (1998). J. Biol. Chem., 273, 30777–30784.

  30. Schaffer EM, Liu JZ, Green J, Dangler CA and Milner JA . (1996). Cancer Lett., 102, 199–204.

  31. Sheen LY, Lii CK, Sheu SF, Meng RH and Tsai SJ . (1996). Food Chem. Toxicol., 34, 971–978.

  32. Shirin H, Pinto JT, Kawabata Y, Soh JW, Delohery T, Moss SF, Murty V, Rivlin RS, Holt PR and Weinstein IB . (2001). Cancer Res., 61, 725–731.

  33. Song JJ and Lee YJ . (2003). Biochem. J., 373, 845–853.

  34. Song JJ, Rhee JG, Suntharalingam M, Walsh SA, Spitz DR and Lee YJ . (2002). J. Biol. Chem., 277, 46566–46575.

  35. Sparnins VL, Barany G and Wattenberg LW . (1988). Carcinogenesis, 9, 131–134.

  36. Srivastava RK, Mi Q, Hardwick JM and Longo DL . (1999). Proc. Natl. Acad. Sci. USA, 96, 3775–3780.

  37. Srivastava RK, Srivastava AR, Korsmeyer SJ, Nesterova M, Cho-Chung YS and Longo DL . (1998). Mol. Cell Biol., 18, 3509–3517.

  38. Sumiyoshi H and Wargovich MJ . (1990). Cancer Res., 50, 5084–5087.

  39. Sundaram SG and Milner JA . (1996). Carcinogenesis, 17, 669–673.

  40. Suzui N, Sugie S, Rahman KM, Ohnishi M, Yoshimi N, Wakabayashi K and Mori H . (1997). Jpn. J. Cancer Res., 88, 705–711.

  41. Takahashi S, Hakoi K, Yada H, Hirose M, Ito N and Fukushima S . (1992). Carcinogenesis, 13, 1513–1518.

  42. Thornberry NA and Lazebnik Y . (1998). Science (Washington, DC), 281, 1312–1316.

  43. Torcia M, De Chiara G, Nencioni L, Ammendola S, Labardi D, Lucibello M, Rosini P, Marlier LNJL, Bonini P, Sbarba PD, Palamara AT, Zambrano N, Russo T, Garaci E and Cozzolino F . (2001). J. Biol. Chem., 276, 39027–39036.

  44. Vantieghem A, Xu Y, Assefa Z, Piette J, Vandenheede JR, Merlevede W, de Witte PAM and Agostinis P . (2002). J. Biol. Chem., 277, 37718–37731.

  45. Wang X, Martindale JL, Liu Y and Holbrook NJ . (1998). Biochem. J., 333, 291–300.

  46. Wargovich MJ . (1987). Carcinogenesis, 8, 487–489.

  47. Wargovich MJ, Woods C, Eng VWS, Stephens LC and Gray K . (1988). Cancer Res., 48, 6872–6875.

  48. Wattenberg LW, Sparnins VL and Barany G . (1989). Cancer Res., 49, 2689–2692.

  49. Xiao D, Pinto JT, Soh J, Deguchi A, Gundersen GG, Palazzo AF, Yoon J, Shirin H and Weinstein IB . (2003a). Cancer Res., 63, 6825–6837.

  50. Xiao D and Singh SV . (2002). Cancer Res., 62, 3615–3619.

  51. Xiao D, Srivastava SK, Lew KL, Zeng Y, Hershberger P, Johnson CS, Trump DL and Singh SV . (2003b). Carcinogenesis, 24, 891–897.

  52. Yamamoto K, Ichijo H and Korsmeyer SJ . (1999). Mol. Cell. Biol., 19, 8469–8478.

  53. You WC, Blot WJ, Chang YS, Ershow A, Yang ZT, An Q, Henderson BE, Fraumeni Jr JF and Wang TG . (1989). J. Natl. Cancer Inst., 81, 162–164.

Download references

Acknowledgements

This investigation was supported in part by USPHS Grants CA101753 and CA55589 (to SVS), CA95191 and CA96989 (to YJL) awarded by the National Cancer Institute. We thank Ajita V Singh, Karen L Lew and Yan Zeng for technical assistance.

Author information

Correspondence to Shivendra V Singh.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Xiao, D., Choi, S., Johnson, D. et al. Diallyl trisulfide-induced apoptosis in human prostate cancer cells involves c-Jun N-terminal kinase and extracellular-signal regulated kinase-mediated phosphorylation of Bcl-2. Oncogene 23, 5594–5606 (2004). https://doi.org/10.1038/sj.onc.1207747

Download citation

Keywords

  • diallyl trisulfide
  • JNK
  • ERK
  • Bcl-2
  • apoptosis
  • prostate cancer

Further reading